LLM prompting involves designing text inputs to generate a response from an LLM. The goal of prompting is to steer the behaviour of the LLM in a way that elicits a desired outcome. Recent research has focused on developing effective prompting techniques that can expand LLMs' capabilities when carrying out a variety of tasks. Examples include prompt patterns [21], in-context instruction learning [22], evolutionary prompt engineering [23] and domain-specific keywords with a trainable gated prompt to guide toward a target domain for general-domain LLMs [24]. Zhong et al. [25] experiment with prompting LLMs to do scientific tasks across fields like business, science, and health by providing the LLM with a research goal and two large corpora, asking the LLM for corpus-level difference. Reppert et al. [26] develop iterated decomposition, a human-in-the-loop workflow for developing and refining compositional LLM programs that improves performance on real-world science question and answer tasks.

services.2Includes respondents working for organizations focused on human resources, legal services, management consulting, market research, R&D, tax preparation, and training. Exhibit 1

Also, responses suggest that companies are now using AI in more parts of the business. Half of respondents say their organizations have adopted AI in two or more business functions, up from less than a third of respondents in 2023 (Exhibit 2)

The biggest increase from 2023 is found in marketing and sales, where reported adoption has more than doubled. Yet across functions, only two use cases, both within marketing and sales, are reported by 15 percent or more of respondents.

“The other thing I want to say is that content does not have to be long form. Gone are the days of creating 5,000 words on this topic to beat your competitor. Needs met doesn’t need 5,000 words. Short-form content works. Short-form answers are good and help the user to find their answer quickly.”